Improved smart city security using a deep maxout network-based intrusion detection system with walrus optimization.

Area of Science:

• Cybersecurity
• Internet of Things (IoT)
• Smart City Infrastructure

Background:

• Smart cities utilize IoT for urban optimization, increasing the need for secure public transportation.
• Securing interconnected digital infrastructure in urban environments is critical.

Purpose of the Study:

• To develop a robust intrusion detection system (IDS) for public transportation in smart cities.
• To address the unique security challenges of IoT-enabled urban transit systems.

Main Methods:

• An IDS model integrating a Deep Maxout Network (DMN) with Walrus Optimization (WO) was developed.
• The DMN-WO model features maxout activation functions for complex pattern recognition in IoT traffic.
• The model is designed for resource efficiency, suitable for real-time deployment on devices like Raspberry Pi.

Main Results:

• The DMN-WO model was trained and validated using CIC-IDS-2018, CIC-IDS-2029 datasets, and real-time data.
• Achieved high performance metrics: 98.06% accuracy, 98.50% detection rate, 98.81% precision, 98.24% specificity, and 98.57% F1-score.
• Demonstrated effectiveness in real-time threat detection within a smart city's public transport network.

Conclusions:

• The research provides a resilient cybersecurity solution for smart city public transportation.
• The DMN-WO model advances threat detection and mitigation in IoT-based urban infrastructure.
• This work establishes a foundation for future real-world deployments in smart city environments.